Air-ground cooperative distribution route optimization method for unmanned aerial vehicle

Abstract

The invention discloses an air-ground cooperative distribution route optimization method for an unmanned aerial vehicle, and belongs to the field of unmanned aerial vehicle system automatic logistics distribution. The method comprises the following steps: firstly, carrying out kinematics and dynamics modeling on an unmanned aerial vehicle in the air-ground collaborative logistics distribution system; secondly, constructing a multi-objective optimization function by taking the distribution distance between each unmanned aerial vehicle and all unmanned vehicles, the flight energy consumption and a safe distance threshold value between the unmanned aerial vehicles as optimization objectives at the same time; obtaining an optimal control signal at the current moment by adopting a particle swarm optimization algorithm; and finally, inputting the optimal control signal into a navigation control system of each unmanned aerial vehicle to generate an optimal distribution route in real time. According to the method, a new thought is provided for unmanned aerial vehicle cluster route planning in an air-ground cooperative distribution scene, the flight energy consumption index item is introduced into traditional unmanned aerial vehicle cluster route planning based on the shortest distribution distance, energy consumption can be saved, and the cruising ability can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of automated logistics distribution for unmanned aerial vehicles, and in particular relates to an air-ground collaborative distribution route optimization method for unmanned aerial vehicles. Background Art

[0002] With the rapid development of e-commerce and the intelligent logistics industry, collaborative delivery technology for unmanned aerial vehicles (UAVs) has become a key breakthrough in improving logistics efficiency. Traditional UAV swarm route planning methods mostly focus on single-objective optimization, primarily designing routes based on the principle of the shortest delivery path. They rarely consider the impact of dynamic energy dissipation on the endurance of UAVs, leading to problems such as low energy utilization and insufficient mission continuity in practical applications. When dealing with air-ground collaborative delivery scenarios, existing technologies often plan paths for UAVs and unmanned vehicles as independent units. This lacks collaborative modeling of the motion coupling characteristics of heterogeneous platforms, making it difficult to achieve global optimal decision-making in dynamic environments. Especially in complex urban scenarios, the safety distance control and multi-objective collaborative optimization issues of UAV swarms are more prominent. Traditional static planning algorithms, due to their poor real-time performance and weak constraint processing capabilities, are prone to route conflicts or obstacle avoidance failures. Summary of the Invention

[0003] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide an air-ground collaborative distribution route optimization method for unmanned aerial vehicles. First, the kinematic and dynamic modeling of the unmanned aerial vehicles involved in the air-ground collaborative logistics distribution system is performed; then, the shortest distance from each unmanned aerial vehicle to all unmanned vehicles to pick up goods and the maximum difference in energy absorption and dissipation are taken as optimization objectives, and the safety distance threshold between unmanned aerial vehicles is used as a penalty term, thereby constructing a multi-objective optimization function; next, the particle swarm optimization algorithm is used to obtain the optimal control signal at the current moment; finally, this optimal control signal is input into the unmanned aerial vehicle cluster navigation control system to generate the optimal distribution route in real time.

[0004] To achieve the above object, the present invention adopts the following technical solutions: A route optimization method for air-ground collaborative delivery of unmanned aerial vehicles is provided. The specific steps of the method are as follows: Step 1: Build kinematic and dynamic models for each UAV; Step 2: Based on the output information of the kinematic model and the real-time location information of the ground unmanned vehicle, construct the delivery distance indicator and the safety distance threshold indicator; Step 3: Construct flight energy consumption indicators based on the output information of the dynamic model; Step 4: Assign weights to the delivery distance index and the flight energy consumption index and add the safety distance threshold index to form a multi-objective optimization function; Step 5: Based on the particle swarm optimization algorithm, calculate the specific values of the multi-objective optimization function corresponding to different particles, and use the particle corresponding to the minimum value as the optimal yaw rate control signal; Step 6: Input the optimal yaw rate control signal into the control system of the unmanned aerial vehicle to complete the air-ground collaborative delivery route optimization task.

[0005] The specific method for building the kinematic model and dynamic model of each unmanned aerial vehicle is as follows: (1) In the northeast coordinate system, The kinematic model of an unmanned aerial vehicle flying in a horizontal plane at a specified altitude is described as follows: in, 、 Respectively represent The horizontal and vertical coordinate values of the unmanned aerial vehicle (the two are collectively referred to as plane position information); and Respectively and The derivative of Indicates the The speed of the unmanned aerial vehicle; and Respectively represent The main purpose of building the kinematic model of the unmanned aerial vehicle is to provide real-time feedback on the heading angle and yaw rate of each sampling moment. The plane position and heading angle information of the unmanned aerial vehicle.

[0006] In order to ensure flight safety, improve stability and protect the actuators of the unmanned aerial vehicle, it is necessary to Limit the speed and roll angle of the unmanned aerial vehicle: Among them, the speed setting value and roll angle setting value Both represent positive real numbers greater than zero.

[0007] In addition, in order to conveniently describe the incident angle of sunlight on the wing surface of the unmanned aerial vehicle, it should also be based on the speed and yaw rate Solve for the roll angle , the specific solution formula is as follows: in, Indicates the acceleration due to gravity, usually 9.8 ; Represents the inverse tangent function.

[0008] (2) Compared with the energy consumption of the signal processing system and the attitude angle adjustment system, the energy consumed by the unmanned aerial vehicle in the low-altitude environment to perform cruising operations while overcoming aerodynamic resistance accounts for the majority. Therefore, the present invention mainly constructs the engine thrust of the unmanned aerial vehicle during level flight. and aerodynamic drag The core dynamic model is:

[0009] in 、 、 They represent the atmospheric density, Oswald efficiency factor, and aspect ratio of the wing respectively; 、 、 、 and Respectively represent Wing area, drag coefficient, zero-lift drag coefficient constant, lift coefficient, and mass of an unmanned aerial vehicle.

[0010] The output information of the kinematic model and the real-time location information of the ground unmanned vehicle are used to construct the delivery distance index item and the safety distance threshold index item. The specific method is as follows: (1) No. The sum of the Euclidean distances between the unmanned aerial vehicle and all ground unmanned delivery vehicles (hereinafter referred to as unmanned vehicles) It can be described by the following mathematical formula: in, 、 Respectively represent The horizontal and vertical coordinates of the unmanned aerial vehicle; 、 Respectively represent The horizontal and vertical coordinate parameters of the unmanned vehicle (the real-time position of the unmanned vehicle); Indicates the total number of unmanned vehicles in the delivery network. However, in actual delivery, we hope that the delivery distance is as short as possible, that is, The smaller the better, which is exactly the opposite of the optimization direction of the optimization function.

[0011] (2) In order to optimize all optimization indicators towards the maximum value, the Euclidean distance can be reciprocated, that is, the delivery distance is defined as (3) Assume that there are a total of To ensure flight safety, the distance between these unmanned aerial vehicles should be greater than or equal to the specified minimum safety distance. Based on this minimum safety distance , the following safety distance threshold indicator items can be constructed: in, Indicates the penalty given when the distance between UAVs is less than the safe distance. This penalty is usually a large negative number, with a recommended value between -1000 and -2000.

[0012] The output information of the dynamic model is used to construct the flight energy consumption index item. The specific method is as follows: (1) According to Article The dynamic model of each unmanned aerial vehicle can be constructed to determine the dynamics of its driving system within a specified time range and the sampling period. Power consumption under : in, represents the efficiency factor of the propeller, represents the integral from time zero to Ts.

[0013] (2) The unmanned aerial vehicle involved in the present invention is able to continuously absorb solar energy during flight because the outer surface of such aircraft is covered with solar panels. The energy absorption power of the solar panels can be constructed in the following steps: 1) Calculate the solar declination angle parameters on the kth day: in, represents the kth day starting from January 1st.

[0014] 2) The hour angle of the sun on day k (unit: rad): in, Represents the specific time in the kth day.

[0015] 3) Calculate the solar altitude angle at that moment (unit: rad): in, Indicates the local latitude (unit: rad).

[0016] 4) Calculate the azimuth of the sun (unit: rad): 5) Calculate the amount of sunlight reaching the The angle of incidence on the solar panels mounted on the wings of an unmanned aerial vehicle (unit: rad): 6) Calculate the direct beam on the horizontal plane of the earth and diffuse reflection : in, represents the direct beam depth, represents the diffuse beam depth, represents a constant coefficient related to solar irradiance, represents the solar irradiance coefficient associated with ground reflection, and G represents the air mass ratio.

[0017] 7) Calculate the Total solar irradiance received by unmanned aerial vehicles : in 、 、 They represent direct beam irradiance, diffuse reflection irradiance, and ground reflection irradiance respectively. Represents the ground reflection coefficient.

[0018] 8) Calculate the energy absorption power of the solar panels covering the fuselage of each unmanned aerial vehicle within a specified time range : in, represents the conversion efficiency coefficient of solar cells, represents the integral from time zero to Ts.

[0019] In summary, flight energy consumption indicators Defined as the energy absorbed by the solar panels covering the UAV body within a specified time The flight energy consumption index items generated during its own flight Difference: The weights of the delivery distance index and the flight energy consumption index are assigned and the safety distance threshold index is added to form a multi-objective optimization function. The specific method is as follows: (1) In order to prevent the optimal solution from being biased towards the optimization target item with a larger order of magnitude, the delivery distance index item and the flight energy consumption index item must be normalized. The processing method of the present invention is to use the following formula to perform normalization after calculating the optimization index item in each rolling time domain: in, and Respectively represent Unmanned aerial vehicle Normalized delivery distance index item and normalized flight energy consumption index item at each sampling moment; and Respectively represent Unmanned aerial vehicle The delivery distance index item and flight energy consumption index item at the sampling moment, Represents the total number of sampling moments selected in each round of iterative optimization.

[0020] (2) Definition Multi-objective optimization function at sampling time : in, Indicates the weight parameter corresponding to the delivery distance indicator item, Indicates the weight parameter corresponding to the flight energy consumption index item.

[0021] The particle swarm optimization algorithm is used to calculate the specific values of the multi-objective optimization function corresponding to different particles, and the particle corresponding to the minimum value is used as the optimal yaw rate control signal. The specific method is as follows. It can be divided into the following 5 steps: (1) According to population size and the predicted time series length , randomly initialized to satisfy the normal distribution OK Particle swarm position matrix of columns and the particle swarm velocity matrix : in It means returning a random number uniformly distributed in the interval (0,1). and Represent the maximum and minimum values of each particle’s position, and Represents the maximum and minimum value of each particle's speed. Particle swarm position matrix Each particle in can be regarded as a yaw rate control signal that needs to be optimized.

[0022] (2) Calculate the particle swarm position matrix Each particle The corresponding multi-objective optimization function value ,in .

[0023] (3) Enter the iterative optimization phase: First, update the position and velocity of the particle swarm; then, limit the particle swarm; finally, calculate the The corresponding multi-objective optimization function value .

[0024] (4) Determine whether the maximum number of iterations has been reached; if so, output the particle corresponding to the minimum value of the current multi-objective optimization function (the optimal yaw angular velocity control signal); otherwise, continue with step (3).

[0025] (5) In the above iterative optimization process, the following weight dynamic attenuation strategy is used to determine the weight parameter of the particle motion speed: : in represents the initial value of the weight, represents the weight attenuation amplitude, represents the total number of iterations, Indicates the current iteration index number.

[0026] Compared with the prior art, the present invention has the following advantages: 1) This paper innovatively proposes a particle swarm optimization (PSO)-based route optimization method for air-ground collaborative delivery in the field of automated logistics distribution for unmanned aerial vehicles (UAVs). By integrating a biological intelligence-based PSO algorithm into the UAV route planning and control framework, it significantly improves global search capabilities and the accuracy of the cost function calculation.

[0027] 2) This invention constructs a multi-objective optimization function, comprehensively considering various optimization objectives, including minimizing the pickup distance between the UAV and the unmanned vehicle, maximizing the difference in energy absorption and dissipation, and a penalty term for the safe distance between the UAVs. This method not only optimizes delivery efficiency but also innovatively incorporates a flight energy consumption indicator, effectively balancing the conflict between delivery efficiency and energy management. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is an overall flow chart of an air-to-ground collaborative delivery route optimization method for unmanned aerial vehicles (UAVs) according to the present invention.

[0029] Figure 2 Flowchart of the yaw rate control signal generation module based on particle swarm optimization.

[0030] Figure 3 This is the route optimization diagram of the air-ground collaborative delivery network under the existing delivery distance constraints.

[0031] Figure 4 This is the route optimization diagram of the air-ground collaborative delivery network under the dual constraints of delivery distance and energy. DETAILED DESCRIPTION

[0032] like Figure 1 As shown in FIG, a route optimization method for air-ground collaborative delivery of unmanned aerial vehicles includes the following steps: (1) Build kinematic and dynamic models for each unmanned aerial vehicle; (2) Based on the output information of the kinematic model and the real-time location information of the ground unmanned vehicle, construct the delivery distance index item and the safety distance threshold index item; (3) Construct flight energy consumption indicators based on the output information of the dynamic model; (4) Assign weights to the delivery distance index and the flight energy consumption index and add the safety distance threshold index to form a multi-objective optimization function; (5) Based on the particle swarm optimization algorithm, the specific values of the multi-objective optimization function corresponding to different particles are calculated, and the particle corresponding to the minimum value is used as the optimal yaw angular velocity control signal; (6) By inputting this optimal yaw rate control signal into the control system of the unmanned aerial vehicle, the air-ground collaborative delivery route optimization task can be completed.

[0033] Example 1. Build kinematic and dynamic models for each unmanned aerial vehicle (1) First, build the following kinematic model of the unmanned aerial vehicle: The speed of each unmanned aerial vehicle is , the total number is 3. The initial position of each unmanned aerial vehicle is set to , , ; The initial heading angle of each UAV is set to , , .

[0034] In order to ensure flight safety, improve stability and protect the actuators of the unmanned aerial vehicle, it is necessary to Limit the speed and roll angle of the unmanned aerial vehicle: in , .

[0035] (2) The present invention mainly constructs the engine thrust of the unmanned aerial vehicle during level flight. and aerodynamic drag The core dynamic model is: The atmospheric density , Oswald efficiency factor , the aspect ratio of the wing , wing area , constant value of zero-lift drag coefficient ,quality .

[0036] 2. Construct delivery distance and safety threshold indicators: (1) No. The sum of the Euclidean distances between the unmanned aerial vehicle and all ground unmanned delivery vehicles (hereinafter referred to as unmanned vehicles) It can be described by the following mathematical formula: Among them, the total number of unmanned vehicles operating in the distribution network The position information of these autonomous vehicles is described by the following time-dependent equation: (2) In order to optimize all optimization indicators towards the maximum value, the Euclidean distance can be reciprocated, that is, the delivery distance is defined as (3) Assume that there are a total of To ensure flight safety, the distance between these unmanned aerial vehicles should be greater than or equal to the specified minimum safety distance. Based on this minimum safety distance , the following safety distance threshold indicator items can be constructed: in, Indicates the penalty for exceeding the safe distance between unmanned aerial vehicles. .

[0037] 3. Construct flight energy consumption indicators (1) Construct the power consumption of each unmanned aerial vehicle Among them, the efficiency factor of the propeller is , sampling period .

[0038] (2) Constructing the energy absorption power of solar panels can be divided into the following steps: Step 1: Study the dynamic evolution of the air-ground collaborative unmanned logistics distribution network at 12 noon on the 202nd day starting from January 1st. Therefore, the declination angle parameter on that day is calculated as in, .

[0039] Step 2: Calculate the hour angle (unit: rad): in, .

[0040] Step 3: Calculate the solar altitude angle at this moment (unit: rad): in, .

[0041] Step 4: Calculate the sun's azimuth (unit: rad): Step 5: Calculate the sun's radiation to the The angle of incidence on the solar panels mounted on the wings of an unmanned aerial vehicle (unit: rad): Step 6: Calculate the direct beam on the horizontal plane of the earth separately and diffuse reflection : The direct beam depth , diffuse beam depth , solar irradiance constant coefficient .

[0042] Step 7: Calculate the Total solar irradiance received by unmanned aerial vehicles : Among them, the ground reflection coefficient .

[0043] Step 8: Calculate the energy absorption power of each UAV's solar panels within a specified time frame : Among them, the conversion efficiency coefficient of solar cells .

[0044] In summary, flight energy consumption indicators Defined as the energy absorbed by the solar panels covering the UAV body within a specified time The flight energy consumption index items generated during its own flight Difference: 4. Constructing a multi-objective optimization function (1) First, design the following normalization formula: in, and Respectively represent Unmanned aerial vehicle Normalized delivery distance index item and normalized flight energy consumption index item at each sampling moment; and Respectively represent Unmanned aerial vehicle The delivery distance index item and flight energy consumption index item at the sampling moment, Represents the total number of sampling moments selected in each round of iterative optimization.

[0045] (2) Definition Multi-objective optimization function at sampling time : in, , .

[0046] 5. Obtain the optimal yaw rate control signal.

[0047] like Figure 2 As shown, it can be divided into the following 5 steps: Step 1: Based on population size =40 and the length of the forecast time series , randomly initialized to satisfy the normal distribution OK Particle swarm position matrix of columns and the particle swarm velocity matrix : Among them, the maximum value of each particle position With minimum value , the maximum value of each particle's speed and minimum value .

[0048] Step 2: Calculate the particle swarm position matrix Each particle The corresponding multi-objective optimization function value ,in .

[0049] Step 3: Enter the iterative optimization phase: First, update the position and velocity of the particle swarm. Then, limit the particle swarm. Finally, calculate the The corresponding multi-objective optimization function value .

[0050] Step 4: Determine whether the maximum number of iterations has been reached; if so, output the particle corresponding to the minimum value of the current multi-objective optimization function (the optimal yaw angular velocity control signal); otherwise, proceed to step (3).

[0051] Step 5: In the above iterative optimization step, the following weight dynamic attenuation strategy is used to determine the weight parameters of the particle motion speed: Among them, the initial value of weight , weight decay amplitude , total number of iterations , Indicates the current iteration index number.

[0052] In order to demonstrate the superiority of the present invention in terms of energy saving, comparative simulations were performed in the examples. Figure 3 The traditional unmanned aerial vehicle swarm delivery route trajectory obtained by optimizing the delivery distance and safety is shown. In addition to the energy consumed by the three unmanned aerial vehicles to provide power for themselves, they also store 1.4663KJ of energy. However, the present invention introduces the flight energy consumption index and designs a reasonable weight between it and the distance index to obtain the following Figure 4 Statistical analysis shows that, after comprehensively considering flight energy consumption and delivery distance, the unmanned aerial vehicle swarm can store 1.5252 kJ of energy within a 50-second delivery time, increasing energy storage efficiency by 4%.

[0053] The present invention deeply integrates air-ground collaborative route planning with the intelligent particle swarm optimization algorithm, and establishes a multi-dimensional collaborative optimization system for the collaborative delivery distance, flight energy consumption, and safety distance threshold of unmanned vehicles and unmanned aerial vehicles. Combined with the efficient optimization characteristics of the particle swarm algorithm, the present invention effectively solves the shortcomings of traditional methods in multi-objective collaboration, dynamic environment adaptation, and real-time control. The technical solution of the present invention not only improves the energy utilization efficiency of the air-ground collaborative logistics distribution system, but also significantly enhances the mission reliability in complex environments through dynamic route optimization, providing an innovative solution for the efficient operation of the air-ground collaborative logistics distribution system.

Claims

1. A route optimization method for air-ground collaborative delivery of unmanned aerial vehicles, characterized by: The steps include: Step 1: Build kinematic and dynamic models for each UAV; Step 2: Based on the output information of the kinematic model and the real-time location information of the ground unmanned vehicle, construct the delivery distance indicator and the safety distance threshold indicator; Step 3: Construct flight energy consumption indicators based on the output information of the dynamic model; Step 4: Assign weights to the delivery distance index and the flight energy consumption index and add the safety distance threshold index to form a multi-objective optimization function; Step 5: Based on the particle swarm optimization algorithm, calculate the specific values of the multi-objective optimization function corresponding to different particles, and use the particle corresponding to the minimum value as the optimal yaw rate control signal; Step 6: Input the optimal yaw rate control signal into the navigation control system of each unmanned aerial vehicle, thus completing the air-ground collaborative delivery route optimization task.

2. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 1, characterized in that: Build the kinematic model and dynamic model of each unmanned aerial vehicle. The specific method is: first, the speed of the i-th unmanned aerial vehicle is and yaw rate As the control input, a kinematic model is constructed that can provide real-time feedback of plane position information, heading angle and roll angle; among them, the roll angle Using the formula Perform an indirect solution, where is the gravitational acceleration constant; then a dynamic model considering engine thrust and aerodynamic drag is constructed.

3. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 2, characterized in that: The kinematic model is as follows: in, 、 Respectively represent The horizontal and vertical coordinates of the unmanned aerial vehicle; and Respectively and The derivative of Indicates the The speed of the unmanned aerial vehicle; and Respectively represent The heading angle and yaw rate of an unmanned aerial vehicle; Consider engine thrust and aerodynamic drag The kinetic model is as follows: in 、 、 They represent the atmospheric density, Oswald efficiency factor, and aspect ratio of the wing respectively; 、 、 、 and Respectively represent Wing area, drag coefficient, zero-lift drag coefficient constant, lift coefficient, and mass of an unmanned aerial vehicle.

4. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 2, wherein: Speed The absolute value of the speed is less than or equal to the speed setting value , roll angle The absolute value of the roll angle is less than or equal to the set value .

5. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 1, characterized in that: The output information of the kinematic model and the real-time position information of the ground unmanned aerial vehicle are used to construct the delivery distance index item and the safety threshold index item. The specific method is as follows: first, the sum of the distances between the i-th unmanned aerial vehicle and the real-time positions of all unmanned aerial vehicles is calculated; then, the inverse of the sum of the distances is used as the delivery distance index item; then, it is determined whether the distance between the unmanned aerial vehicles exceeds a given threshold. If so, a negative number with an absolute value in the range of 1000-2000 is designed as the safety distance threshold index item.

6. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 1, characterized in that: The output information of the dynamic model is used to construct a flight energy consumption index. The specific method is as follows: first, the power consumption of the drive system of each unmanned aerial vehicle is calculated; then, the energy absorption power of the solar panels covering the fuselage of each unmanned aerial vehicle is calculated; and finally, the difference between the energy absorption power and the energy consumption power of each unmanned aerial vehicle is used as the flight energy consumption index.

7. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 1, characterized in that: Assign weights to the delivery distance index and the flight energy consumption index and add the safety distance threshold index to form a multi-objective optimization function. The specific method is as follows: first calculate the Sampling moments and their future totals After normalizing the delivery distance index item and the flight energy consumption index item corresponding to the moment, Represents the length of the predicted time series; then, a weight parameter with a sum of 1 is assigned to the normalized delivery distance index item and the flight energy consumption index item, and the safety distance threshold index item is added to form the final multi-objective optimization function.

8. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 7, characterized in that: Calculate the Sampling moments and their future totals The method for normalizing the delivery distance index item and the flight energy consumption index item corresponding to the time is: Normalized delivery distance index item ,in Indicates the Unmanned aerial vehicle Delivery distance index item at each sampling moment; Normalized flight energy consumption index item ,in Indicates the Unmanned aerial vehicle Flight energy consumption index items at each sampling moment.

9. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 1, characterized in that: Based on the particle swarm optimization algorithm, the specific values of the multi-objective optimization function corresponding to different particles are calculated, and the particle corresponding to the minimum value is used as the optimal yaw rate control signal. The specific method is as follows: first, a particle swarm is randomly generated according to the population size and the length of the predicted time series. Next, a weighted dynamic attenuation strategy is used to iteratively optimize the particle swarm. Finally, the particle corresponding to the minimum value of the multi-objective optimization function at the current moment is used as the optimal yaw rate control signal for a certain unmanned aerial vehicle.

10. The method for optimizing air-ground collaborative delivery routes for unmanned aerial vehicles according to claim 9, characterized in that: Weight dynamic attenuation strategy, its specific calculation formula is: ,in, The weight parameter representing the particle speed, represents the initial value of the weight, represents the weight attenuation amplitude, represents the total number of iterations, Indicates the current iteration index number.

Images